Not at all. The -taub- option fires up a rank correlation and that
doesn't depend on either variable being binary. In addition,
there are several other possibilities here, such as those
mentioned by Ron�n Conroy. Ron�n didn't say where the programs
he used come from, but -findit- will identify locations.
Deidra Young
I have a sample size of 315 cases distributed evenly between Australia and
an International database. 55 cases have autism. All cases have another
genetic condition. In this case I am looking at the ordered distribution of
the disturbed awake breathing rhythm (thee levels from none, mild & severe).
I am also doing logistic (where autism is the DV and binary) and ologit
(where I use the ordered categorical severity items as DV) so that I can
adjust for age. However, I wanted to learn how to do tabs without using the
problematic chi squared test which is inappropriate for ordered, categorical
measures and cannot inform which are the cells being compared.
Is this method only appropriate for comparing two groups (df = 1)?
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
